exa-x-search

name	exa-x-search
description	Search tweets and X/Twitter content using Exa. Finds social discussions, product announcements, developer opinions, trending topics, and community sentiment. Use when searching for tweets, X/Twitter discussions, or social media sentiment.
user-invocable	true

X/Twitter Search (Exa)

Tool

Call the native MCP tool exa-search__web_search_advanced_exa directly with category: "tweet".

Always include category: "tweet".

Critical Filter Restrictions

The tweet category has very limited filter support. These params cause errors:

Param	Status
includeText	NOT SUPPORTED — 400 error
excludeText	NOT SUPPORTED — 400 error
includeDomains	NOT SUPPORTED — 400 error
excludeDomains	NOT SUPPORTED — 400 error
moderation	NOT SUPPORTED — 500 error

Use the query string itself to filter content — put keywords, hashtags, and exclusions in query.

Supported Parameters

• query, numResults, type
• Date: startPublishedDate, endPublishedDate, startCrawlDate, endCrawlDate
• Content: textMaxCharacters, enableHighlights, enableSummary
• additionalQueries — for hashtag variations
• livecrawl=preferred — for recent tweets

Call Examples

Tool: exa-search__web_search_advanced_exa

# Recent tweets on a topic
{ query: "Claude Code MCP experience", category: "tweet", startPublishedDate: "2025-01-01",
  numResults: 20, type: "auto", livecrawl: "preferred" }

# Product announcements (keywords in query, not includeText)
{ query: "announcing new open source AI tool release", category: "tweet",
  startPublishedDate: "2025-12-01", numResults: 15, type: "auto" }

# Developer sentiment
{ query: "developer experience frustrating painful bug", category: "tweet",
  numResults: 20, type: "deep", livecrawl: "preferred" }

# Competitor monitoring
{ query: "[INCUMBENT_PRODUCT] review feedback", category: "tweet",
  startPublishedDate: "2025-06-01", numResults: 25, type: "auto" }

When to Use

• Social discussions and community sentiment
• Product announcements from company accounts
• Developer opinions and experiences
• Trending topics
• Expert takes and threads
• Competitor/market signal monitoring

Output Format

Return:

1. Results — tweet content, author handle, date, engagement if visible
2. Sources — tweet URLs
3. Notes — sentiment summary, notable accounts, threads vs single tweets

Note: tweet content can be informal, sarcastic, or context-dependent.